Hi, my name is
Pratham Agrawal.
I work for the Data.
'You have to dream before your dreams can come true.' ~ Dr. APJ Abdul Kalam. I overhead this line, which keeps me hustling. .
About Me
I'm a software developer who turns data into stories. 📊✨
I craft scalable, efficient systems that transform raw information into actionable insights. From optimizing query performance for real-time dashboards to building robust data pipelines that handle petabytes of data, I love the challenge of turning complex problems into elegant solutions.
My toolkit includes Python, RabbitMQ, Docker, and PostgreSQL, and I'm always eager to learn new technologies. Whether I'm architecting a resilient archival service or developing an innovative clustering algorithm, I strive for both performance and user-friendliness.
Beyond coding, I'm a strong believer in collaboration and clear communication. I enjoy working with teams to simplify processes, ensure data integrity, and ultimately empower businesses with smarter, faster tools.
Let's build something awesome together – trust me, it'll be a wild ride! 🚀
- Python
- C/C++
- SQL
- Git
- RabbitMQ
- Kafka
- Apache Airflow
- Spark
- Hadoop
- Tableau
- PostgreSql
- Oracle
- MongoDB
- Database Warehousing
- Database Design
Where I’ve Worked
Software Developer @ Affisnys
January 2024 - Presents
- Analytics Service: Led the maintenance of an Analytics service that processes and delivers data for 20+ dashboards and 200+ charts, empowering real-time insights while optimizing user customization features. Optimized query performance and API response by implementing materialized SQL views, reducing query overload by 40%. Developed an Analytics Scheduler to automate the delivery of 100+ analytics reports across multiple services, sending recurring emails to numerous recipients and ensuring timely dissemination of insights.
- Archival Service: Developed and implemented a robust archiving service to manage service-level data across 10+ services with varying foreign key relationships, ensuring referential integrity and seamless data consistency. The service includes a failover mechanism, safeguarding data during migration and maintaining reliability in case of system failures.
- Eventlogger Service: Developed an efficient service to process pub/sub events with async database writes, consolidating data from 10+ services into 40+ tables and simplifying schema management.
- Collaborated with the DevOps team to manage production-level data and service migrations for 5+ clients, ensuring data integrity and minimizing downtime.
Some Things I’ve Built
Featured Project
LogHive
A robust, scalable Python logging library that enables distributed log collection with advanced connection management, automatic reconnection, and thread-safe logging capabilities. 🌟
- Python
- RabbitMQ
- Redis
- PostgresDB
- Docker
![LogHive]()
Featured Project
Clusterbase
Clusterbase is a system that is easily configurable on the number and type of parameters and clustering rules, where in the clustering happens in real time with the values dynamically provided by the user. The Web application provides an user with a console to add a new cluster along with options to view and query all the active clusters.
- Python
- NextJS
- Google Firebase
![Clusterbase]()
Featured Project
Crypto2csv
A python command line interface for retrieving crytocurrency data in a readable csv format. Covered 80+ parameters over a wide range of cryptocurrencies, with various data fetching options like intervals, coins, etc.
- Python
- Command Line Utility
- API data extraction
![Crypto2csv]()
Featured Project
Sportlight
Provided a solution for highlights extraction for sport games using Machine learning and APIs, thereby reducing the time of watching a match by 90 %. Formulated the NLTK model using wordnet, and NLP which tested out to be 93% accurate
- React
- Python
- NLTK
- NLP
- Javascript
- CSS
![Sportlight]()
Featured Project
Olympics Data Visulization
Constructed a dataset comprising 2.7 lakh rows information of about Athletes who participated in the Olympics.Developed a web application to illustrate the trends and outliers for 124 years of Olympics History using Python, Dash and Plotly.
- Python
- R Programming
- Plotly
- Dash
- Flask
![Olympics Data Visulization]()
Other Projects
Image Colorisation
A tool to colorise and enchance images using Deep neural Networks.
IPL Analysis
A solution for to answer complex data analytical questions or take data-driven decisions, or complex data visualizations for IPL Analysis.
LeetCode Solutions
This repository consists of some of the solutions of the questions I did on LeetCode.
Publications
Computational Approaches for Prediction of Cardiovascular Risks along with API Feed
The research consists of 2 parts: Namely, (a) cardiovascular ailments prediction using Machine Learning/ Neural Networks Algorithms. (b) Predicting if a user is healthy enough based on fitness tracker data.
The Quest for a Martian Life
An Outlook for comparing the different views if terraforming Mars should be acheived or saving the planet.
Artificial Intelligence at Healthcare Industry
The Article contains insights on how AI based technology has helped the medical sector. The article also shows a comparison between the past, present and the future of Artificial Intelligence in the medical sector.
What’s Next?
Get In Touch
I'm looking for new opportunities. My inbox is always open. I will try my best to get back to you.
Say Hello